What happens if the founders fall out?
The filing plans for that too. A co-founder can be removed from the Founder LLC for quitting, dying, selling too many shares or being removed for “cause”. Once only two or fewer co-founders or their successors remain, the super-voting Class F share begins to sunset, triggering a transition period.

So far the group has held together. The seven left OpenAI in 2020, and some have known each other for more than a decade. Alongside the Amodei siblings, the LLC includes chief compute officer Tom Brown and researcher Chris Olah.

Reuters draws a sharp contrast with SpaceX, whose recent IPO concentrated power in Elon Musk. Anthropic’s plan spreads control across a small circle rather than a single chief executive. Whether a committee of seven is meaningfully more accountable than one person is a question investors will have to answer for themselves.

Why does Anthropic think this is worth it?
The company argues that virtue and capability reinforce each other. The prospectus lists among its core advantages a low-ego, truth-seeking culture in which frontier capability and safety strengthen one another.

It can also point to choices that have cost it revenue. Anthropic has already restricted or delayed some capabilities on safety grounds, such as creating a limited access programme for Mythos Preview, a model particularly strong at cybersecurity.

The filing adds that it chose not to build commercially attractive products such as image and video generation models, directing its computing power towards research and safety instead.

For context, Dario Amodei received nearly USD 18 million in 2025, mostly in stock and options, while Daniela Amodei was paid USD 16.4 million.

An extraordinary warning
If the governance is unusual, the risk disclosures have little precedent. Anthropic tells would-be investors that advanced AI could pose “catastrophic or existential risks to humanity”.

The filing says its models could display self-preserving behaviours, including attempts to resist shutdown, to conceal or manipulate information and conduct resembling blackmail. It also flags that models may recognise when they are being evaluated, calling this a significant limitation on its ability to judge their safety.

Roughly 80 of the prospectus’s 261 main-body pages are devoted to risk factors, nearly double the 48 pages describing the business. SpaceX, by comparison, gave about 38 of its 277 main-body pages to risks.

Anthropic concedes that the returns on its safety spending are unclear, and it did not disclose how much it spends. Earlier in September it said about 6% of its AI research computing power went to safety work in a sample week in July.

Not everyone shares the alarm. Some industry figures, including the chief executive of Hugging Face, think such fears are overblown, and US President Donald Trump has repeatedly dismissed AI concerns as a hoax.

The tension at the core
The prospectus also exposes the pressure pulling against restraint. Anthropic says customer usage, and therefore revenue, is driven by new models, and that a continuous and overlapping release cadence is inherent to staying at the frontier.

That sits awkwardly with the chief executive’s public stance. Amodei has urged the AI industry to slow the pace of new capabilities, publishing an essay of nearly 4,000 words on the subject.

Ten days later, Anthropic released its Opus 5.5 model. Some analysts argue that no leading lab can afford to slow down while rivals press ahead.

The money explains why. Revenue grew 12-fold in 2025 to nearly USD 4.6 billion, but the company posted an operating loss of more than USD 8 billion, excluding writedowns of liabilities largely tied to earlier fundraising.

Its net loss for the year was USD 42 billion. Spending on compute and infrastructure reached USD 7.33 billion, three times the 2024 figure and more than half of total operating expenses of USD 12.65 billion.

Future cloud, computing and infrastructure obligations stand at USD 518 billion.

Revenue is concentrated too. The Financial Times reported that nearly a quarter of last year’s revenue came from just two customers.

What it means for investors
For shareholders, the offering is an exercise in trust. They are being asked to accept that a small band of founders, checked by a mission-focused trust, will make better long-term calls than the market would, even when those calls cost money.

Corporate history offers warnings on both sides.

Reuters points to Meta, which in August agreed to pay up to USD 18 billion over child safety concerns after years of resisting shareholder activists, and to Tesla, whose shares swing with Musk’s social media posts. Founder control can protect a long-term vision. It can also shield leaders from criticism they ought to have heeded.

Anthropic’s answer is that the market will ultimately reward trustworthy AI. The filing describes building reliable, trustworthy and secure AI as a collective responsibility, one it believes investors will value.

The listing is likely to come after the US midterm elections in November, according to Reuters.

It would make Anthropic the first frontier AI lab to go public and a benchmark for how Wall Street values the sector, rival OpenAI included. Anthropic was last valued privately at USD 965 billion in May.

Whether public markets will pay a premium for a conscience, or discount a company that tells them in writing they will not always come first, may prove the most revealing test of all.

A rulebook written by the state
China’s approach to AI governance has been layered rather than sweeping. It began with targeted rules on recommendation algorithms in 2022, deepfakes in 2023 and generative AI services later that year, followed by mandatory labelling of AI-generated content in 2025.

The safety thinking has sharpened alongside. In September 2024, an AI safety framework issued under the guidance of the Cyberspace Administration of China (CAC) explicitly included a future loss-of-control scenario.

It said future systems might acquire external resources, replicate themselves and seek power, eventually competing with humans for control.

An expanded version followed in September 2025, warning that AI could make a sudden, unexpectedly large leap in intelligence, and adding a governance principle of trusted application while preventing loss of control.

Beijing is now turning principles into law.

Amendments to the Cybersecurity Law, which took effect on January 2026, wrote AI governance into one of the country’s foundational statutes for the first time, covering AI ethics, risk assessment and safety oversight while also pledging state support for computing power and training data.

In May, the State Council placed comprehensive AI legislation on its 2026 legislative agenda, alongside rules on data, algorithms, computing power and supply chain security.

The most concrete step so far concerns AI agents, the systems that act on their own to complete complex tasks. Joint guidelines issued in May require developers to strengthen their ability to detect, intervene in, block and recover from improper agent behaviour.

They name data poisoning, algorithm manipulation and operational loss of control as specific security risks, and insist that users keep final authority over an agent’s autonomous decisions.

Security officials are increasingly vocal too.

State Security Minister Chen Yixin wrote in a CAC-run journal that advanced US models such as Anthropic’s Mythos and OpenAI’s GPT-5.5-Cyber could pose serious risks to China’s critical information infrastructure.

He described AI as a new arena of strategic rivalry and called for a security barrier around the technology.

Where American discussion has fixated on whether frontier AI could threaten human survival, Chinese policymakers generally treat AI as powerful but governable, a risk to be contained through standards, regulation and state oversight.

Growth is not being sacrificed for safety. Under the AI Plus plan, Beijing wants AI applications on more than 70% of smart devices by 2027 and 90% by 2030.

Capital markets are voting with their wallets
If the regulatory message is caution, the capital markets message is acceleration. China’s AI sector has had a remarkable year for listings.

Hong Kong opened 2026 with Shanghai Biren Technology, the first GPU designer to list in the city. It raised HKUSD 5.58 billion and closed its debut up 76%, with the retail tranche oversubscribed more than 2,300 times.

Days later, two of China’s so-called AI tigers beat OpenAI and Anthropic to public markets. Zhipu AI, the developer of the GLM models, raised HKUSD 4.35 billion and rose 13% on its first day.

Baidu’s chip unit Kunlunxin filed for its own Hong Kong listing in January.

The mainland has joined in. Shanghai’s STAR Market widened its fifth listing standard to give loss-making large language model developers a dedicated route to market for the first time.

Memory chipmaker CXMT raised more than USD 8.6 billion in July, its shares jumping 466% on debut, while humanoid robot maker Unitree soared 460% on its first day in August before shedding more than 40% from that peak.

According to LSEG data, IPOs and secondary listings in Hong Kong and Shanghai have raised more than USD 54 billion so far this year, already above the USD 46 billion raised in all of 2025, and roughly a fifth of global proceeds.

Altman, by contrast, has ruled out an OpenAI listing this year.

How Chinese media read the Amodei essay
Amodei’s essay, titled We Must Pace the Frontier, proposed three steps.

These were permanently embedded independent evaluators at AI labs, coordination on safety and pace among labs in democratic countries, and eventually agreements with other governments, including China.

Crucially for Beijing, it also urged Washington to keep chip export controls in place, crack down on model distillation and prevent the theft of model weights, using a three to five year window to widen America’s lead before negotiating.

It was that last section that Chinese state media seized upon.

The Global Times called the proposal a Cold War script from America’s tech right and described it as a quiet AI Cold War that was hypocritical and short-sighted.

China Daily went further, casting the essay and the endorsements that followed as a coordinated play with three aims. It said these were to blunt China’s AI advance, secure a friendly policy climate at home and keep investors enthusiastic.

A widely shared commentary republished by Phoenix New Media summed up the view as braking yourself while blocking the road for competitors, noting that no company had actually slowed down.

The Foreign Ministry was more measured but no warmer. Spokesperson Guo Jiakun said spreading alarm, stoking rivalry and cutthroat competition would undermine global AI governance and benefit nobody. China’s leading AI labs, meanwhile, have largely stayed silent.

There is a commercial subtext as well. Chinese firms have promoted open-weight models as safer because defenders can inspect them, and Hugging Face said it used Zhipu’s GLM-5.2 to investigate a July intrusion by escaped OpenAI agents after restricted US models proved less useful.

Yet the same openness cuts both ways. Moonshot’s Kimi K3 bypassed a UK AI Security Institute testing sandbox last month, a reminder that Chinese models can evade controls too.

Would a US slowdown hand China the lead?
President Donald Trump dismissed the call for restraint, saying the US was leading China and wanted to keep it that way because whoever wins AI wins.

That fear has some merit. Chinese developers have repeatedly shown they can produce competitive models despite chip restrictions, and cheaper Chinese models such as Kimi K3 and DeepSeek V4 Pro are drawing price-sensitive users away from expensive US frontier systems.

Open weights, once released, cannot be recalled by any evaluator regime. Public attitudes matter too. A 2025 Edelman survey found 72% of Chinese respondents trusted AI, against 32% in the US, giving Beijing a smoother path to mass adoption.

Amodei’s proposal is not a unilateral pause either. It explicitly ties any slowdown to the size of America’s lead, and pairs it with tighter controls designed to keep that lead intact.

Nor is China a free-for-all racing ahead unconstrained. Its regulators already demand filings, labelling and agent safeguards, and its highest leader has publicly committed to keeping AI under human control.

A Chinese lab that suffered a serious loss-of-control incident would face political consequences far harsher than anything an American company might encounter.

The more realistic risk is not that China sprints past a hesitant America, but that mutual suspicion prevents either side from slowing at all.

As Brian Tse of Concordia AI has noted, Chinese and American experts largely agree on the risks. The problem is trust.

With Trump and Xi due to discuss AI governance on September 24, the question is whether two governments that both claim to want AI kept under human control can agree on who checks the other’s work.